Deep learning to discover and predict dynamics on an inertial manifold

Abstract

A data-driven framework is developed to represent chaotic dynamics on an inertial manifold (IM), and applied to solutions of the Kuramoto-Sivashinsky equation. A hybrid method combining linear and nonlinear (neural-network) dimension reduction transforms between coordinates in the full state space and on the IM. Additional neural networks predict time-evolution on the IM. The formalism accounts for translation invariance and energy conservation, and substantially outperforms linear dimension reduction, reproducing very well key dynamic and statistical features of the attractor.

arXiv commentAccepted in Physical Review E

References in corpus (11)

• Deep learning for universal linear embeddings of nonlinear dynamics
• Data-driven discovery of coordinates and governing equations
• Discovering physical concepts with neural networks
• Data-Driven Forecasting of High-Dimensional Chaotic Systems with Long Short-Term Memory Networks
• Data-assisted reduced-order modeling of extreme events in complex dynamical systems
• Construction of Reduced Order Models for Fluid Flows Using Deep Feedforward Neural Networks
• Revealing the state space of turbulent pipe flow by symmetry reduction
• Hyperbolicity and the effective dimension of spatially-extended dissipative systems
• Reduction of SO(2) symmetry for spatially extended dynamical systems
• Estimating dimension of inertial manifold from unstable periodic orbits
• Periodic orbit analysis of a system with continuous symmetry - a tutorial

Cited by in corpus (31)

• Super-Resolution Analysis via Machine Learning: A Survey for Fluid Flows
• Data-driven discovery of intrinsic dynamics
• Learning hydrodynamic equations for active matter from particle simulations and experiments
• Data-Driven Reduced-Order Modeling of Spatiotemporal Chaos with Neural Ordinary Differential Equations
• OnsagerNet: Learning Stable and Interpretable Dynamics using a Generalized Onsager Principle
• Nonlinear proper orthogonal decomposition for convection-dominated flows
• Revealing the state space of turbulence using machine learning
• Data-driven low-dimensional dynamic model of Kolmogorov flow
• Data-driven control of spatiotemporal chaos with reduced-order neural ODE-based models and reinforcement learning
• Reinforcement-learning-based control of convectively-unstable flows
• Machine learning based non-Newtonian fluid model with molecular fidelity
• Symmetry-reduced Dynamic Mode Decomposition of Near-wall Turbulence
• Deep learning delay coordinate dynamics for chaotic attractors from partial observable data
• Data-driven state-space and Koopman operator models of coherent state dynamics on invariant manifolds
• Exact coherent structures in two-dimensional turbulence identified with convolutional autoencoders
• Generalizability of reservoir computing for flux-driven two-dimensional convection
• Slow Invariant Manifolds of Singularly Perturbed Systems via Physics-Informed Machine Learning
• Data-driven modeling of the regular and chaotic dynamics of an inverted flag from experiments
• Learning normal form autoencoders for data-driven discovery of universal,parameter-dependent governing equations
• Global and Local Reduced Models for Interacting, Heterogeneous Agents
• Dynamics of a Data-Driven Low-Dimensional Model of Turbulent Minimal Pipe Flow
• Learning a Reduced Basis of Dynamical Systems using an Autoencoder
• Real-time forecasting of chaotic dynamics from sparse data and autoencoders
• Stochastic Latent Transformer: Efficient Modelling of Stochastically Forced Zonal Jets
• Elastoinertial turbulence: Data-driven reduced-order model based on manifold dynamics
• Prospects of federated machine learning in fluid dynamics
• Particles to Partial Differential Equations Parsimoniously
• Data-driven guessing and gluing of unstable periodic orbits
• Using machine-learning modelling to understand macroscopic dynamics in a system of coupled maps
• Learning emergent PDEs in a learned emergent space
• Latent-Lagrangian Neural Networks for Reduced Order Modeling of Non-autonomous Nonlinear Dynamical Systems